Proteases and degradation of extracellular matrix in neurodegeneration

Matrix Metalloproteinase-9 as an Important Contributor to the Pathophysiology of Depression

Matrix metalloproteinases (MMPs) are physiologically expressed in the central nervous system in neurons, astrocytes and microglia, and their aberrant elevation contributes to a number of diseases. Amongst the MMP members, MMP−9 has generated considerable attention because of its possible involvement in inflammatory responses, blood-brain barrier permeability, the regulation of perineuronal nets, demyelination, and synaptic long-term potentiation. Emerging evidence indicate an association between MMP−9 and the syndrome of depression. This review provides an updated and comprehensive summary of the probable roles of MMP−9 in depression with an emphasis on the mechanisms and potential of MMP−9 as a biomarker of depression.

Persistent motor deficit following infusion of autologous blood into the periventricular region of neonatal rats

Periventricular hemorrhage (PVH) in the brain of premature infants is often associated with developmental delay and persistent motor deficits. Our goal is to develop a rodent model that mimics the behavioral phenotype. We hypothesized that autologous blood infusion into the periventricular germinal matrix region of neonatal rats would lead to immediate and long-term behavioral changes. Tail blood or saline was infused into the periventricular region of 1-day-old rats. Magnetic resonance (MR) imaging was used to demonstrate the hematoma. Rats with blood infusion, as well as saline and intact controls, underwent behavior tests until 10 weeks age. Blood-infused rats displayed significant delay in motor development (ambulation, righting response, and negative geotaxis) to 22 days of age. As young adults, they exhibited impaired ability to stay on a rotating rod and to reach for food pellets. MR imaging at 10 weeks demonstrated subsets of rats with normal appearing brains, focal cortical infarcts, or mild hydrocephalus. There was a good correlation between MR imaging and histological findings. Some rats exhibited periventricular heterotopia and/or subtle striatal abnormalities not apparent on MR images. We conclude that autologous blood infusion into the brain of neonatal rats successfully models some aspects of periventricular hemorrhage that occurs after premature birth in humans.

Does thrombin play a role in the pathogenesis of brain damage after periventricular hemorrhage?

Neonatal periventricular hemorrhage (PVH) is a devastating complication of prematurity in the human infant. Based upon observations made primarily in adult rodents and the fact that the immature brain uses proteolytic systems for cell migration and growth, we hypothesized that thrombin and plasmin enzyme activities contribute to the brain damage after PVH. The viability of mixed brain cells derived from newborn rat periventricular region was suppressed by whole blood and thrombin, but not plasmin. Following injection of autologous blood into the periventricular region of newborn rat brain, proteolytic activity was detected in a halo around the hematoma using membrane overlays impregnated with thrombin and plasmin fluorogenic substrates. Two-day old rats received periventricular injection of blood, thrombin, and plasminogen. After 2 days, thrombin and blood were associated with significantly greater damage than saline or plasminogen. Two-day old mice received intracerebral injections of blood in combination with saline or the proteolytic inhibitors hirudin, alpha2macroglobulin, or plasminogen activator inhibitor-1. After 2 days, hirudin significantly reduced brain cell death and inflammation. Two-day-old mice then received low and high doses of hirudin mixed with blood after which behavioral testing was conducted repeatedly. At 10 weeks there was no statistically significant evidence for behavioral or structural brain protection. These results indicate that thrombin likely plays a role in neonatal periventricular brain damage following PVH. However, additional factors are likely important in the recovery from this result.

Injections of blood, thrombin, and plasminogen more severely damage neonatal mouse brain than mature mouse brain

The mechanism of brain cell injury associated with intracerebral hemorrhage may be in part related to proteolytic enzymes in blood, some of which are also functional in the developing brain. We hypothesized that there would be an age-dependent brain response following intracerebral injection of blood, thrombin, and plasminogen. Mice at 3 ages (neonatal, 10-day-old, and young adult) received autologous blood (15, 25, and 50 microl respectively), thrombin (3, 5, and 10 units respectively), plasminogen (0.03, 0.05, and 0.1 units respectively) (the doses expected in same volume blood), or saline injection into lateral striatum. Forty-eight hours later they were perfusion fixed. Hematoxylin and eosin, lectin histochemistry, Fluoro-Jade, and TUNEL staining were used to quantify changes related to the hemorrhagic lesion. Damage volume, dying neurons, neutrophils, and microglial reaction were significantly greater following injections of blood, plasminogen, and thrombin compared to saline in all three ages of mice. Plasminogen and thrombin associated brain damage was greatest in neonatal mice and, in that group unlike the other 2, greater than the damage caused by whole blood. These results suggest that the neonatal brain is relatively more sensitive to proteolytic plasma enzymes than the mature brain.

Endogenous proteolytic activity in a rat model of spontaneous cerebral stroke

We evaluated the expression of two extra-cellular protease systems in a model of spontaneous cerebrovascular pathology: spontaneously hypertensive stroke-prone rats (SHRSP). The appearance of brain damage in individual animals was imaged and followed by means of magnetic resonance imaging (MRI). In situ zymography of brain slices obtained 3 days after the appearance of brain damage showed an increase in plasminogen activator (PA)/plasmin activity that co-localised with the cerebral damage detected by MRI; there was also concomitant accumulation/activation of inflammatory cells in the damaged area. Proteolytic activity was inhibited by the urokinase-specific inhibitor amiloride but not by an antibody against tissue-type plasminogen activator (t-PA). SDS-PAGE zymography of brain extracts revealed the presence of 58 kDa plasminogen-dependent lysis areas in the ischemic and non-ischemic tissues, and a 33 kDa lysis area in ischemic tissue only. An antibody against t-PA inhibited the former, whereas the latter was inhibited by amiloride. The specific induction of urokinase-type plasminogen activator (u-PA) in the damaged tissue was further confirmed by the fact that both u-PA protein mass and mRNA were markedly increased in the damaged cerebral areas. Concomitant metalloproteinase-2 (MMP-2) activation was only observed in the damaged area. These data suggest that u-PA is expressed and selectively catalyses proteolysis in the injured area of spontaneous brain damage in SHRSP.

Acute tissue damage after injections of thrombin and plasmin into rat striatum

BACKGROUND AND PURPOSE

Extravasation of blood is associated with intracerebral hemorrhage and head trauma. The mechanism of brain cell injury associated with hemorrhage differs from that due to pure ischemia. The purpose of this study was to investigate the acute changes after intracerebral injections of proteins that are involved in blood clotting and clot lysis.

METHODS

Sixty-eight adult rats were subjected to stereotaxic intrastriatal injections of normal saline (5 microL), low- (2.5 U/5 microL) and high-dose (25 U/5 microL) thrombin, low- (0.1 microgram/5 microL) and high-dose (1 microgram/5 microL) tissue plasminogen activator, low- (0.05 U/5 microL) and high-dose (0.5 U/5 microL) plasminogen, and low- (0.335 U/5 microL) and high-dose (3.35 U/5 microL) plasmin. Forty-eight hours later rats were perfusion fixed. Brain damage area, eosinophilic neurons, terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate-biotin nick end labeling (TUNEL)-positive cells, infiltrating neutrophils, CD8a immunoreactive leukocytes, and reactive microglia were quantified.

RESULTS

Damage area in striatum, dying cells, inflammatory cells, and microglial reaction were significantly greater after the high-dose plasminogen, plasmin, and thrombin injections. Tissue plasminogen activator injections were associated with mild inflammation.

CONCLUSIONS

These results suggested that thrombin and plasmin are harmful to brain cells in vivo. Although the doses required to cause damage are relatively great in consideration of the plasma content of these proteins, their pathological effect might be enhanced through synergism with other mechanisms.

Control elements between -9.5 and -3.0 kb in the human tissue-type plasminogen activator gene promoter direct spatial and inducible expression to the murine brain

Tissue-type plasminogen activator (t-PA) participates in the control of synaptic plasticity and memory formation in the central nervous system (CNS). Transgenic mice harbouring either 9.5, 3.0 or 1.4 kb of the human t-PA promoter fused to the LacZ reporter gene were used to assess t-PA promoter-directed expression in vivo. The 9.5 kb t-PA promoter directed expression to the brain, most notably to the dentate gyrus, superior colliculus, hippocampus, thalamus and piriform cortex. Staining was also observed in the retrosplenial and somatosensory cortex. The 3.0 kb t-PA promoter directed generalized and poorly defined expression to the cortex and hippocampus, while the 1.4 kb t-PA promoter directed expression selectively to the medial habenula. Intravenous administration of lipopolysaccharide into mice harbouring the 9.5 kb t-PA promoter resulted in an increase in reporter gene activity in the lateral orbital cortex and thalamus. Results of in vitro transfection experiments of NT2 cells with a series of t-PA promoter deletion constructs confirmed the presence of regulatory elements throughout the 9.5 kb promoter region. Finally, we describe a cis-acting element related to the NFAT recognition site that provides a protein-binding site and which may play a role in the selective expression of the 1.4 t-PA promoter in the medial habenula. These results indicate that elements between -3.0 and -9.5 kb of the t-PA promoter confer constitutive and inducible expression to specific regions of the CNS.

Dying neural cells activate glia through the release of a protease product

The close relationship between neurodegeneration and gliosis could play a relevant role in propagating the degenerative event in the brain. Although there is evidence of the neurotoxicity of activated glia, the ability of damaged neurons to modulate glial response remains unexplored. Exposure of primary glial cells to damaged or dead hippocampal neurons was followed by glial release of tumor necrosis factor-alpha (TNF-alpha). This release was reduced by a partial prevention of neural death. By contrast, no TNF-alpha was released when glial cells were exposed to damaged murine fibroblasts. Exposure of glial cells to the cerebrospinal fluid (CSF) of patients with Alzheimer's disease was also followed by TNF-alpha release, while the CSF of subjects with nondegenerative brain disorders evoked no response. These data suggest that damaged neurons both in vitro and in vivo release factor(s) that activate glial response. Heat treatment of sonicated neurons or use of a mixture of protease inhibitors, among them the caspase inhibitors Z-DEVD-FMK and Z-YVAD-FMK, prevented TNF-alpha release from glial cells. We conclude that a primary neurodegenerative event may induce glial response by releasing a neurospecific protein factor via activation of a caspase.